Research On Indoor Positioning Method Of Location Fingerprint Based On Quadtree Algorithm

With the rapid development of Internet technology,more and more people pay attention to the new business field based on location,which promotes the rapid development of location technology,especially indoor location technology.Due to the advantages of low cost and wide coverage of WiFi and other wireless networks,the traditional location fingerprint locating method based on Received Signal Strength Indicator(RSSI)has become an ideal locating method.However,due to various factors,the traditional location fingerprint algorithm has the problem of large computational load or low positioning accuracy.Thesis improves the WKNN algorithm and proposes a fingerprint location method based on quadtree search and fractal direction entropy weighting(QSFDEW).The main research work is as follows:1?Several typical and common matching algorithms and the technical basis of fingerprint location method are studied,and the factors affecting the complex propagation of WiFi signal in indoor are systematically analyzed.The influence of WiFi signal is studied from the aspects of acquisition times,different time,device distance,and device orientation,and a four-direction multiple acquisition scheme is determined.The collected RSSI data is subjected to limit-mean double filtering,and after filtering out abnormal data,the original location fingerprint database is constructed.2?In order to improve positioning efficiency and accuracy,a positioning algorithm based on QSFDEW is proposed in thesis.The algorithm creates the original fingerprint database,then divides the experimental area by quadtree algorithm and stores the collected data in grid format to create a new fingerprint database.In the positioning stage,a quadtree index is performed to quickly search the adjacent quadrants to determine the smallest sub-region to which the point to be located belongs.3?After determining the minimum sub-region to which the point to be located belongs,combined with the idea of entropy weighting in different directions,the K reference points closest to the point to be located are screened in different directions,and different entropy weights are given to these K points.Determine the final estimated position information of the point to be located.The QSFDEW algorithm proposed in thesis has been verified by experiments.The experiments show that the average positioning error of QSFDEW is 0.99 m and the average positioning time is 0.31 s.Compared with the Bayesian,KNN and WKNN algorithms,the positioning accuracy and efficiency are improved.At the same time,compared with several newer methods at present,it also has better positioning accuracy and achieves the expected effect.